Health Sciences and Medicine Research Methods - Essay Example

Health Sciences and Medicine Research Methods Research Methods Question a) What is meant by the terms clinical heterogeneity and statistical heterogeneity? (4) Heterogeneity refers to the variability or differences between studies in the estimates of effects of healthcare interventions. Clinical heterogeneity is the difference between studies in the key characteristics of the participants, interventions or outcome measures. Statistical heterogeneity is the differences in the reported effects (Bland, 2004). b) Comment on the clinical heterogeneity of the trials in Figure 1. (2) Clinical heterogeneity exists when patient – level factors, that is variables related to patient characteristics, cormobidites and accompanying treatment may influence or modify the magnitude of the treatment effect. In the clinical trials of Treatment A and Placebo clinical heterogeneity is present and leads to a clinically relevant impact on the treatment. c) Comment on the statistical heterogeneity of the trials in Figure 1. (4) The test for heterogeneity indicate a Chi ² = 6.25, df= 5(P-0.2) I²= 20% There is low heterogeneity due to the low P value of 0.2 and an I² of 20%. This means there is little difference or variation in statistical heterogeneity between the studies in the estimates of the effects of treatment A and placebo. d) Which study provides the greatest weight in the meta-analysis and why? (4) The statistical heterogeneity tests carry more weight than the clinical heterogeneity because the overall effect shows a Z value of 1.69 and a low value of 0.09. e) What is the overall pooled effect and associated 95% confidence interval for the meta-analysis? Is the pooled result in Figure 1 statistically significant? (4) f) Define publication bias. (2) Publication bias occurs when a research is published where the findings of the research were not representative of the population of the completed study. The results of the study differ from results of all research that have been done in that area. The readers are in danger of drawing the wrong conclusion which can be dangerous if the treatment is said to be safe and effective in the study (Donner, Piaggio, & Villar, 2001). g) Do you think publication bias could be present in the meta-analysis in Figure 1? Justify your answer. (4) I do not think there is any publication bias but it would be difficult to conclude with the limited information provided only by Figure 1. Question 2 a) When and why was the Declaration of Helsinki created? (1) The Declaration of Helsinki is a set of the ethical principles for medical research involving human subjects adopted by the 18th World Medical Association (WMA) general assembly in Helsinki, Finland in June 1964. b) Comment on whether research without advance consent is ever justified. (2) Research without advance consent is never justified. This is because of ethical considerations and the protection of the rights of the persons being used in the research being violated. The Declaration of Geneva of the World Medical Association binds the physician with the words, ‘‘The health of my patient will be my first consideration,’’ and the International Code of Medical Ethics declares that, ‘‘A physician shall act only in the patient’s interest when providing medical care which might have the effect of weakening the physical and mental condition of the patient.’’ In medical research on human subjects, considerations related to the well-being of the human subject should take precedence over the interests of science and society. c) When recruiting participants to a study, you should supply them with a Participant Information Sheet to aid informed consent. What information should it provide? (4) The partcipation in a research should be as a result of a choice made by the participants. The researcher will require their consent. The participant informatiom is a sheet that aids the partipant to make this choice (Edwards, Braunholtz, Lilford, & Stevens, 1999). The Partcipant Information Sheet should provide the following; 1) The language used should be understood by a 12 year old. 2) Information on the purpose of the study, background, procedures involved, expected duration of the study and what will happen to any blood or tissue samples. 3) A description of any possible risks to the participant that might arise during and/or after the study 4) Description of the benefit to the participant 5) Specific steps that will be taken regarding confidentiality d) Define: i) De-identified data. This is data in which identifying iinformation has been removed. This includes the patient’s name, medical record, birth date, social security number in order to protect the privacy of the patient. ii) Anonymous data. This is data that can never be reasonable connected to the person providing them. Example questionnaires returned by mail with no return address. iii) Pseudonymous data. This is data that can be associated with a particular person. Question 3 a) Define a case-control study. (2) This is a form of observational study where two differing existing groups in outcome are compared based on common characteristics. They are mainly used in studies to determine what causes a particuler medical condition.One group will have the condition while the other one will not have it.It is normally used in epidimeology studies. b) Provide a research question where a case-control study is an appropriate design. Justify your choice. (4) The research question one would ask is if the participant drinks heavily in a study to determine the causes of liver cirrhosis. This is because alcohol consumption is normally associated with causing liver cirrhosis (Egger, Davey-Smith, Schneider, & Minder, 1997). c) Identify two potential biases within a case-control study. (4) Information bias which results from systematic differences in the way the data on exposure or outcome was collected. For example the participants may have been asked for how long they have had the condition. This information is dependent on the patient’s ability to recall how long they have had the condition. Selection bias occurs when there is a systematic difference between those in the treatment part of the study and those in the control group. For example if people are asked to volunteer for a study in most cases the volunteers will be persons who are health conscious but would this be representative of the general population (World Medical Association, 2001). Question 4 a) What is meant by the term ‘cluster randomized controlled trial’? (2) A cluster randomized controlled trial is a trial where groups of participants are randomized as opposed to individuals. b) Provide an example of where a cluster RCT would be appropriate. Justify your choice.(4) A community in the rural area may be affected by an endemic disease such as Malaria. The whole village would be randomised to have the intervention rather than the individuals in that community. It would be prudent to administer the vaccine to the group as it is more convenient and economical. Question 5 A trial of scale and polish for the treatment of periodontal (gum) disease in pregnant women was undertaken to assess the effect on preterm births. The results were as follows: Preterm birth Full term Scale and polish 10 297 307 No intervention 15 285 300 25 582 590 a) What is meant by the terms risk ratio and odds ratio? (4) The risk ratio is the probability that a member of an exposed group will develop a disease relative to the probability that a member of an unexposed group will develop the same disease. Odds ratio is the odds of the disease among the exposed individual divided by the odds of the disease among the unexposed. b) Calculate the risk ratio for the data presented (show your calculations). (4) Risk Ratio = P (Pre-Term/ scale and polished) / P (Pre-term / no intervention) = (10/307) / (15/300) = 0.033/0.05 = 0.66 c) Calculate the odds ratio for the data presented (show your calculations). (4) Odds Ratio = P (Pre term/scale & polish) / 1 - P (Pre term/scale & polish) P ( Pre term/no intervention / 1 - P (Pre term/no intervention) = 10/307 / 1- 10/307 = 0.033 / 0.967 = 0.34 / 0.05 = 6.8 15/300 / 1- 15/300 0.05 / 0.95 d) When would you expect the risk ratio and odds ratio to differ? (2) You would expect differences in the risks and odds ratio if there is a high statistical heterogeneity in the samples. e) What is meant by number need to treat? (2) This is he number of patients that need to be treated in order for one to benefit. It provides a simple summary of a clinical trial with one digit. f) Calculate the NNT for the data presented (show your calculations). (4) NNT = 1/ Absolute Risk Reduction = 1 (297/307) – (285/300) = 1 / 0.96-0.95 = 1/ 0.01= 100 g) The confidence interval around the risk is (0.39 to 1.78). Comment on the implications of the findings of the trial. (5) It means that 100 pregnant women need to be treated with scale and polish for gum disease in order to avoid pre term delivery. Question 6 a) Choose a health related topic of interest to you. Develop the aims, objectives and hypotheses (where appropriate) for your chosen topic, stating an appropriate study design. (8) Tooth Decay Aim of the Study; To determine the preventions of tooth decay in children Objectives of the study: To examine whether the use of flouride based toothpaste reduces tooth decay in children Study Design. The study design that will be used for the study will be observational. There will be two groups of children; one with tooth decay and the other without. The groups will be selected from patients of dentists from a particular health clinic. b) Provide a brief outline (500-800 words) of the methods for the proposed study, including details of the population of interest, the interventions or exposures of interest, any relevant comparisons to be made and the relevant outcomes to be measured. Discuss steps you might take to minimize bias within your chosen study design (17). Tooth decay is one of the major childhood diseases affecting children aged 5 – 8 years old. Toddlers are exposed to a lot of sugary foods that cause tooth decay. The study’s aim is to determine prevention measures of tooth decay in toddlers. It will involve the use of fluoride based toothpaste as the intervention. Tooth decay results in dental cavities and is caused when bacteria in the mouth make acids that eat away at a tooth. Tooth decay can cause pain, infection and eventually tooth loss. The combination of the food we eat and bacteria causes tooth decay. Plaque is always forming on our teeth and gums and contains bacteria. This in turn feed on the sugars that we eat and forms acid which attacks the teeth and causes tooth decay. Children are more prone to tooth decay as they love to eat sugary foods. Many children develop cavities and this indicates tooth decay. Research has shown that tooth decay as one of the leading childhood diseases in toddlers with children having tooth decay by the time they are 5 years old. The study will be conducted using the observational study design. This was selected as opposed to the experimental studies that are more expensive. The observational study will ensure validity of the results, efficiency and consider the practical and ethical considerations. The observation study will look into the prevention of tooth decay in toddlers. The population for the study will be children aged 3-6 years old. There will be two groups of participants, those that use the toothpaste and those that do not. The selection of the participants will ensure that the children selected will already have tooth decay but with no toothache. The control group will have no tooth decay at all. Selection bias is where there is a difference between the children with tooth decay and those that are in the control group. This type of bias may come about due to determining the level of tooth decay that the children have. To avoid this kind of bias, participants will be drawn from patients of particular dentists who will ensure that the level of tooth decay is the same. The children will not come from the general population. This has the advantage of it being easy to identify the participants and it will be less expensive as they will come from one health institution. This will make it easy for the characteristics of the cases and those of the control to be comparable. The recall exposure for both groups will be the same. The intervention that will be used for the study is a particular type of fluoride toothpaste. This is a preventive measure that would reduce the occurrence of tooth decay in children. The study will make comparisons with other available treatments. Tooth decay is can also be treated by a local anesthetic injection. When the area is numb, the decay is drilled away before filing the cavity. Tooth decay can also be treated by sealing the decay into the teeth with filling materials or under a crown. This method does not involve any injections. References Bland, M., 2004. BMeta – Analysis: Heterogeneity and Publication Bias. New York: University of York. Donner, A., Piaggio, G., and Villar, J., 2001. Statistical methods for the meta-analysis of cluster randomization trials. Statistical Methods in Medical Research, 10, pp. 325-338. Edwards, S. J. L., Braunholtz, D. A., Lilford, R. J., and Stevens, A. J., 1999. Ethical issues in the design and conduct of cluster randomized controlled trials. BMJ, 318, pp. 1407-1409. Egger, M., Davey-Smith, G., Schneider, M., and Minder, C., 1997. Bias in meta-analysis detected by a simple, graphical test. British Medical Journal, 315, pp. 629–634. World Medical Association, 2001. Declaration of Helsinki. Bulletin of World Health Organisation. Read More